The clinical study CAO-/ARO-/AIO-94 of the German Rectal Cancer Study Group led clinicians around the world to embrace neoadjuvant chemoradiotherapy as the treatment of choice for patients with locally advanced rectal cancer because it reduces the rate of local recurrence. The patients that we analyzed for our initial gene expression analyses for response prediction4 were participants in that study. A follow-up prospective randomized phase III clinical trial is now in place (CAO-/ARO-/AIO-04). The goal of this trial is to address whether tumor response and prognosis can be further improved by the addition of oxaliplatin to the existing combined modality therapy (5-FU and radiation). The gene expression analyses, aCGH experiments, and measurement of miRNA profiles that we propose in Project 3 will be done on patients enrolled in this trial. Our collaborations with Gttingen are integrated with a Clinical Research Unit with the title The biological basis of individual tumor response in patients with rectal cancer (http://www.kfo179.de/). The establishment of such Clinical Research Units in Germany is highly competitive and is aimed at establishing a climate and an infrastructure conducive to interdisciplinary research that is focused on clinically relevant questions. The Clinical Research Units consists of eight subprojects. We are actively involved in Subproject 1, Gene expression signature and genetic polymorphisms for response prediction of rectal carcinomas to preoperative chemoradiotherapy and Subproject 2, Functional validation of genes involved in resistance of rectal carcinomas to preoperative chemoradiation. The other subprojects analyze other pertinent aspects of rectal cancer biology and relevant clinical parameters, including the detection of micrometastases, the development of an anti-CEA-based immuno-PET/CT in vivo imaging system for therapy monitoring, the analysis of mismatch-repair status as it relates to therapy response, and the identification of biomarkers (including specific SNPs in genes in drug metabolizing pathways) to assess treatment-related toxicity. Together, we are able to pursue a holistic approach to the problem of treatment failure and disease prognostication in patients with rectal cancer. There is a wide spectrum of tumor responsiveness of rectal adenocarcinomas to preoperative chemoradiotherapy ranging from complete response to complete resistance. We were therefore interested to investigate whether parallel gene expression profiling of the primary tumor can contribute to stratification of patients into groups of responders or non-responders. We have collected pre-therapeutic biopsies from 30 locally advanced rectal carcinomas (determined by rectal ultrasound as uT3 and uT4) and analyzed these samples for gene expression signatures using microarrays. All patients were participants of a phase III clinical trial (CAO/ARO/AIO-94, German Rectal Cancer Trial) and were randomized to receive a preoperative combined modality therapy including 5-fluorouracil and radiation. In an initial set of 23 patients responders and non-responders (measured by T-level down-sizing) showed significantly different expression levels for 54 genes (p less than 0.001). The ability to predict response to therapy using gene expression profiles was rigorously evaluated using leave-one-out cross-validation (LOOCV). Tumor behavior was correctly predicted in 83% of patients (p=0.02). Sensitivity (correct prediction of response) was 78% and specificity (correct prediction of non-response) was 86% with a positive predictive value of 78% and negative predictive value of 86%. These results suggest that pre-therapeutic gene expression profiling may assist in response prediction of rectal adenocarcinomas to preoperative chemoradiotherapy. The implementation of gene expression profiles for treatment stratification and clinical management of cancer patients requires validation in large, independent studies, which are now under way. In parallel to the validation of the 54 gene set in Gttingen we plan to profile all 300 rectal cancer samples (and matched normal mucosa) using the Agilent 44K expression platform at the NCI, a number required for a test robust enough to enter clinical practice. To date, 107 tumor samples and matched mucosa have already been successfully processed. Obviously, these analyses not only provide insight into mechanisms of response to chemoradiotherapy but also into basic pathways disturbed in rectal carcinomas and are thus thematically linked to Project 1 and 2. We consider the availability of matched normal mucosa as a definitive enrichment of the dataset. Depending on the genes that we identify as differentially expressed, we will consider whether DNA sequence analysis might be appropriate to identify potential polymorphisms associated with protein function. We hypothesize that modification of expression levels of genes involved in resistance to chemoradiotherapy could be explored for the purpose of sensitizing a priori resistant tumors to treatment. For instance, silencing of genes that are upregulated in resistant tumors in cell lines established from these tumor samples could potentially result in increased sensitivity. This line of research will be pursued in Gttingen as subproject 2 of the Clinical Research Unit. Dr. Marian Grade, who was a postdoctoral fellow in my laboratory, was awarded a Junior Group Leader position to focus on the functional validation of response genes. Due to our expertise we will establish cell lines from resistant and responsive tumors in my laboratory for functional analysis and will continue to collaborate with Dr. Grade. These analyses will be complemented by establishing microRNA expression profiles from responsive and non-responsive patients. Since one function of miRNA is the regulation of gene expression, we anticipate that differences in the expression profiles could indeed reveal biomarkers of response. In addition, we have begun to identify cell lines derived from primary colorectal carcinomas that show differential sensitivity to 5-FU and radiation. We anticipate that differential gene expression profiles after treatment can help to identify genes involved in resistance which can then be validated in primary tumors.